In this paper we address the task of learning how to segment a particular class of objects, by means of a training set of images and their segmentations. In particular we propose a method to overcome the extremely high training time of a previously proposed solution to this problem, Kernelized Structural Support Vector Machines. We employ a one-class SVM working with joint kernels to robustly learn significant support vectors (representative image-mask pairs) and accordingly weight them to build a suitable energy function for the graph cut framework. We report results obtained on two public datasets and a comparison of training times on different training set sizes.

Learning Graph Cut Energy Functions for Image Segmentation / Manfredi, Marco; Grana, Costantino; Cucchiara, Rita. - ELETTRONICO. - (2014), pp. 960-965. (Intervento presentato al convegno 22nd International Conference on Pattern Recognition, ICPR 2014 tenutosi a Stockholm, Sweden nel Aug. 24-28) [10.1109/ICPR.2014.175].

Learning Graph Cut Energy Functions for Image Segmentation

MANFREDI, MARCO;GRANA, Costantino;CUCCHIARA, Rita
2014

Abstract

In this paper we address the task of learning how to segment a particular class of objects, by means of a training set of images and their segmentations. In particular we propose a method to overcome the extremely high training time of a previously proposed solution to this problem, Kernelized Structural Support Vector Machines. We employ a one-class SVM working with joint kernels to robustly learn significant support vectors (representative image-mask pairs) and accordingly weight them to build a suitable energy function for the graph cut framework. We report results obtained on two public datasets and a comparison of training times on different training set sizes.
2014
22nd International Conference on Pattern Recognition, ICPR 2014
Stockholm, Sweden
Aug. 24-28
960
965
Manfredi, Marco; Grana, Costantino; Cucchiara, Rita
Learning Graph Cut Energy Functions for Image Segmentation / Manfredi, Marco; Grana, Costantino; Cucchiara, Rita. - ELETTRONICO. - (2014), pp. 960-965. (Intervento presentato al convegno 22nd International Conference on Pattern Recognition, ICPR 2014 tenutosi a Stockholm, Sweden nel Aug. 24-28) [10.1109/ICPR.2014.175].
File in questo prodotto:
File Dimensione Formato  
2014ICPR.pdf

Open access

Tipologia: Versione originale dell'autore proposta per la pubblicazione
Dimensione 1.15 MB
Formato Adobe PDF
1.15 MB Adobe PDF Visualizza/Apri
Pubblicazioni consigliate

Licenza Creative Commons
I metadati presenti in IRIS UNIMORE sono rilasciati con licenza Creative Commons CC0 1.0 Universal, mentre i file delle pubblicazioni sono rilasciati con licenza Attribuzione 4.0 Internazionale (CC BY 4.0), salvo diversa indicazione.
In caso di violazione di copyright, contattare Supporto Iris

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11380/1060473
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 1
  • ???jsp.display-item.citation.isi??? 0
social impact